AAL atlas

Access

See nilearn.datasets.fetch_atlas_aal.

Notes

This Deterministic atlas is the result of an automated anatomical parcellation of the spatially normalized single-subject high-resolution T1 volume provided by the Montreal Neurological Institute (MNI) (Collins et al.[1]).

Using this parcellation method, three procedures to perform the automated anatomical labeling of functional studies are proposed: (1) labeling of an extremum defined by a set of coordinates, (2) percentage of voxels belonging to each of the AVOI intersected by a sphere centered by a set of coordinates, and (3) percentage of voxels belonging to each of the AVOI intersected by an activated cluster.

For the AAL version SPM 5, 8, and 12, the map image (data.maps) contains 117 unique integer values that define the parcellation. However, these values are not consecutive integers from 0 to 116, as is usually the case in Nilearn. Therefore, they should not be interpreted as indices for the list of label names.

In contrast, the total number of parcellations in AAL 3v2 is 167. The 3v2 atlas contains 171 unique integer values that define the parcellation. These values are consecutive integers from 0 to 170, except for the anterior cingulate cortex (35, 36) and thalamus (81, 82), which are left empty in AAL 3v2. In addition, the region IDs are provided as strings, so it is necessary to cast them to integers when indexing.

For example, with version SPM 5, 8 and 12, to get the name of the region corresponding to the region ID 5021 in the image, you should do:

# This should print 'Lingual_L'
data.labels[data.indices.index("5021")]

Conversely, to get the region ID corresponding to the label “Precentral_L”, you should do:

# This should print '2001'
data.indices[data.labels.index("Precentral_L")]

For more information on this dataset’s structure, see https://www.gin.cnrs.fr/en/tools/aal/, AAL template for SPM 12[2], and Tzourio-Mazoyer et al.[3].

Content

• atlas_type: str. Type of atlas. See Probabilistic atlas and Deterministic atlas.

• indices: list of str. Indices mapping labels to values in the maps image. This list has 117 elements in version SPM 5, 8 and 12, and 167 elements in version 3v2. Since the values in the ‘maps’ image do not correspond to indices in labels, but rather to values in indices, the location of a label in the labels list does not necessary match the associated value in the image. Use the indices list to identify the appropriate image value for a given label.

• labels: list of str. List of the names of the regions. There are 117 names in version SPM 5, 8, and 12, and 167 names in version 3v2.

• lut: pandas.DataFrame. Act as a look up table (lut) with at least columns ‘index’ and ‘name’. Formatted according to ‘dseg.tsv’ format from BIDS.

• maps: str. Full path to 3D nifti file containing label image. The image has shape (91, 109, 91) and contains 117 unique integer values defining the parcellation in version SPM 5, 8 and 12, and 167 unique integer values defining the parcellation in version 3v2.

• template: str. The standardized space of analysis in which the data is provided.

References

[1]

D.L. Collins, A.P. Zijdenbos, V. Kollokian, J.G. Sled, N.J. Kabani, C.J. Holmes, and A.C. Evans. Design and construction of a realistic digital brain phantom. IEEE Transactions on Medical Imaging, 17(3):463–468, 1998. doi:10.1109/42.712135.

[2]

Aal template for spm 12. https://www.gin.cnrs.fr/en/tools/aal/. Accessed: 2021-05-19.

[3]

N. Tzourio-Mazoyer, B. Landeau, D. Papathanassiou, F. Crivello, O. Etard, N. Delcroix, B. Mazoyer, and M. Joliot. Automated anatomical labeling of activations in spm using a macroscopic anatomical parcellation of the mni mri single-subject brain. NeuroImage, 15(1):273–289, 2002. doi:10.1006/nimg.2001.0978.

License

unknown